Pinjia He
14 papers · 2021–2026 · 6 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+5 more ↓ Show less ↑
🏃 Academic Marathon · Sprint 🌉 Interdisciplinary Bridge · Silver 🗺️ Taxonomy Completionist (28) 🌍 Conference Polyglot · Silver 🤝 Dynamic Duo (10)
🗃️
Keyword Collector
(73)
❓
The Questioner
(2)
⚡
Prolific Year · Productive
🌈
Renaissance Researcher
(5)
💎
Century Club · Bronze
Top co-authors
Research topics
Artificial Intelligence
(13)
Core AI · Artificial Intelligence
(12)
Natural Language Processing
(6)
Resources & Methods
(5)
Deep Learning
(4)
Machine Learning
(4)
Learning Types · Machine Learning
(3)
Learning Types · Deep Learning
(3)
Computer Science
(2)
Applications · Computer Science
(2)
Applications · Natural Language Processing
(2)
Computer Vision
(1)
Knowledge & Reasoning
(1)
Core AI · Computer Vision
(1)
Learning Paradigms
(1)
Keywords
large language model
(9)
benchmark evaluation
(4)
multimodal large language model
(3)
ai safety
(2)
visual question answering
(2)
software testing
(2)
prompt engineering
(1)
in-context learning
(1)
logical reasoning
(1)
model safety
(1)
adversarial defense
(1)
automated reasoning
(1)
harmful content
(1)
safety alignment
(1)
code generation
(1)
adversarial attack
(1)
software engineering
(1)
multimodal learning
(1)
static analysis
(1)
model alignment
(1)
Papers
Curing Miracle Steps in LLM Mathematical Reasoning with Rubric Rewards
ACL 2026
MicLog: Towards Accurate and Efficient LLM-based Log Parsing via Progressive Meta In-Context Learning
AAAI 2026
SHAPE: Unifying Safety, Helpfulness and Pedagogy for Educational LLMs
ACL 2026
Can’t See the Forest for the Trees: Benchmarking Multimodal Safety Awareness for Multimodal LLMs
ACL 2025
Insight Over Sight: Exploring the Vision-Knowledge Conflicts in Multimodal LLMs
ACL 2025
OpenRCA: Can Large Language Models Locate the Root Cause of Software Failures?
ICLR 2025
Refuse Whenever You Feel Unsafe: Improving Safety in LLMs via Decoupled Refusal Training
ACL 2025
ToolSafety: A Comprehensive Dataset for Enhancing Safety in LLM-Based Agent Tool Invocations
EMNLP 2025
UTBoost: Rigorous Evaluation of Coding Agents on SWE-Bench
ACL 2025
Difficult Task Yes but Simple Task No: Unveiling the Laziness in Multimodal LLMs
EMNLP 2024
Does ChatGPT Know That It Does Not Know? Evaluating the Black-Box Calibration of ChatGPT
COLING 2024
GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher
ICLR 2024
LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models
EMNLP 2024
SANRAZOR: Reducing Redundant Sanitizer Checks in C/C++ Programs
OSDI 2021